Pharmacokinetics and Pharmacodynamics of Antibody-Drug Conjugates Administered via Subcutaneous and Intratumoral Routes

Abstract

We hypothesize that different routes of administration may lead to altered pharmacokinetics/pharmacodynamics (PK/PD) behavior of antibody-drug conjugates (ADCs) and may help to improve their therapeutic index. To evaluate this hypothesis, here we performed PK/PD evaluation for an ADC administered via subcutaneous (SC) and intratumoral (IT) routes. Trastuzumab-vc-MMAE was used as the model ADC, and NCI-N87 tumor-bearing xenografts were used as the animal model. The PK of multiple ADC analytes in plasma and tumors, and the in vivo efficacy of ADC, after IV, SC, and IT administration were evaluated. A semi-mechanistic PK/PD model was developed to characterize all the PK/PD data simultaneously. In addition, local toxicity of SC-administered ADC was investigated in immunocompetent and immunodeficient mice. Intratumoral administration was found to significantly increase tumor exposure and anti-tumor activity of ADC. The PK/PD model suggested that the IT route may provide the same efficacy as the IV route at an increased dosing interval and reduced dose level. SC administration of ADC led to local toxicity and reduced efficacy, suggesting difficulty in switching from IV to SC route for some ADCs. As such, this manuscript provides unprecedented insight into the PK/PD behavior of ADCs after IT and SC administration and paves the way for clinical evaluation of these routes.

Keywords: antibody-drug conjugate (ADC); intratumoral administration; local toxicity; modeling and simulation (M&S); monomethyl auristatin E (MMAE); pharmacokinetics/pharmacodynamics (PK/PD); subcutaneous administration.

Figure 1

An overview of the experimental design. (a) PK study following IT, SC, and IV administration of 10 mg/kg T-vc-MMAE in NCI-N87 tumor-bearing nude mice; (b) efficacy study of IT, SC, and IV administered T-vc-MMAE (high, mid, and low doses) in NCI-N87 tumor-bearing nude mice; (c) local toxicity study of SC administered T-vc-MMAE, naked mAb (trastuzumab), naked payload (MMAE), and vehicle control in wild-type and nude mice.

Figure 2

Schematic diagram of the semi-mechanistic PK/PD model developed for IV, IT, and SC administered T-vc-MMAE ADC. The PK model consists of the plasma and tumor PK models connected via vascular (extravasation) and surface exchange (diffusion). The PK model is connected to the PD model using intracellular target engagement (%TE) predicted by the PK model, which is used to drive the efficacy of T-vc-MMAE. Please refer to the PK and PD model structure sections in the method section for a detailed description of the symbols and disposition processes captured by the model.

Figure 3

Observed plasma pharmacokinetics (PK) of ADC analytes in mice after intravenous (IV), intratumoral (IT), and subcutaneous (SC) administration of 10 mg/kg of T-vc-MMAE single dose. The figure displays the mean (SD) observed concentrations of: (a) total antibody; (b) total MMAE; (c) unconjugated MMAE; and (d) conjugated MMAE in plasma.

Figure 4

Observed tumor pharmacokinetics (PK) of ADC analytes in mice after intravenous (IV), intratumoral (IT), and subcutaneous (SC) administration of 10 mg/kg of T-vc-MMAE single dose. The figure displays the mean (SD) observed concentration of: (a) total antibody; (b) total MMAE; (c) unconjugated MMAE; and (d) conjugated MMAE in the tumor.

Figure 5

In vivo efficacy of T-vc-MMAE ADC after intravenous (IV), intratumoral (IT), and subcutaneous (SC) administration. The figures show the mean (SD) tumor growth curves (upper) and individual tumor growth curves from each animal (lower) after IT (10, 3, 1 mg/kg), SC (20, 6, 2 mg/kg), and IV (10, 3, 1 mg/kg) administration of T-vc-MMAE single dose, along with the untreated group.

Figure 6

Local toxicity of subcutaneously administered ADC in wild-type mice. Representative histopathology in wild-type mice that received: (a) 30 mg/kg of T-vc-MMAE single-dose subcutaneously, (b) 0.5 mg/kg of MMAE single dose subcutaneously, (c) 30 mg/kg of trastuzumab single dose subcutaneously, and (d) vehicle control subcutaneously. The microscope magnification was 20×. Tissue damage is indicated by arrows displaying necrosis (black), ulceration (red), inflammatory infiltrate (blue), hyperkeratosis (green), blisters (cyan), acanthosis (brown), and hypergranulosis (orange). The tissue damage shown includes necrosis (a1), ulceration (a2), inflammatory infiltrate (a3,a4) are shown. The represented tissue damage of hypergranulosis (b1), acanthosis (b2), inflammatory infiltrate (b3) are shown. (e) The histogram displays total tissue damage scores (mean and SD) calculated by summation of severity scores (0~3) of blister formation, inflammatory infiltrate, ulceration, and necrosis; * p < 0.05, ** p = 0.005, *** p = 0.0005, **** p < 0.0001.

Figure 7

Comparison of model fitted and observed PK profiles of T-vc-MMAE analytes in plasma and tumors after intravenous (IV), intratumoral (IT), and subcutaneous (SC) administration of 10 mg/kg of T-vc-MMAE single dose. The figure displays observed (dots) and model-predicted (solid lines) plasma and tumor concentration vs. time profiles of total antibody (red), total MMAE (black), and unconjugated MMAE (cyan) in mice.

Figure 8

Comparison of PK/PD model fitted and observed tumor growth curves after treatment with T-vc-MMAE administered via intravenous (IV), intratumoral (IT), and subcutaneous (SC) routes. The figure displays observed (red dots) and model-predicted (solid lines) tumor growth curves after treatment with high (red bar), mid (green bar), and low (blue bar) ADC doses.

Figure 9

PK/PD model simulated tumor growth inhibition (TGI) after treatment with T-vc-MMAE administered via intravenous (IV), intratumoral (IT), or subcutaneous (SC) routes. (a) Simulation for clinically approved dosing regimens of 1.8 mg/kg given Q3W for 6 cycles; (b) simulation of dosing amounts required to achieve similar TGI after IV, IT, and SC administration with the same dosing frequency (Q3W); (c) simulation of dosing frequency to achieve similar TGI after IV, IT, and SC administration with the same dosing amount (1.8 mg/kg).

Figure 9

PK/PD model simulated tumor growth inhibition (TGI) after treatment with T-vc-MMAE administered via intravenous (IV), intratumoral (IT), or subcutaneous (SC) routes. (a) Simulation for clinically approved dosing regimens of 1.8 mg/kg given Q3W for 6 cycles; (b) simulation of dosing amounts required to achieve similar TGI after IV, IT, and SC administration with the same dosing frequency (Q3W); (c) simulation of dosing frequency to achieve similar TGI after IV, IT, and SC administration with the same dosing amount (1.8 mg/kg).